Italian Semester of Presidency of the European Union

EUROPEAN CONFERENCE OF MINERVA

Quality for cultural Web sites
Online Cultural Heritage for Research, Education and Cultural Tourism Communities

Parma, 20-21 November 2003, Auditorium Paganini

Theodore S. Papatheodorou
(High Performance Information Systems Laboratory University of PatrasGreece)

Dimitris K. Tsolis
(High Performance Information Systems LaboratoryUniversity of PatrasGreece)

Vito Cappellini
(Excellence Centre for Communication and Media Integration, University of Florence Italy)

Alessandro Piva
(National Inter-university Consortium for TelecommunicationsUniversity of Florence Italy)

Technologies for IPR and data protection

Introduction

Due to the large diffusion of the Internet, multimedia systems in general are getting more and more importance and a new and unusual way to distribute some information, such as cultural content for this specific case, is offered. The chance to be able to look at works of art in an interactive manner has led to the birth of many virtual museums and art galleries all over the web; moreover the owners of image databases and the managers of Cultural Heritage are deeply interested in the possibility of spreading their goods to the larger number of worldwide users.
At the same time, IPR protection of cultural content is gradually becoming a critical issue primarily for the following reasons:

1. Advances in technology have improved the ability to reproduce, distribute, manage and publish information: reproduction costs are much lower for both legitimate IPR holders (content owners) and those infringing intellectual property legislation, and digital copies are perfect replicas. Computer networks have changed the economics of distribution: networks enable the distribution of multimedia content worldwide, cheaply and quickly. As a consequence, it is easier and less expensive both for the rights holder to distribute a work and for an individual to make and distribute unauthorized copies. Finally, the World Wide Web has fundamentally altered the publication of information, allowing everyone to be a publisher with worldwide reach.
   
   
2. The production and sharing of information in electronic form has been integrated into everyday life, directly affecting intellectual property legislation. Today, casual everyday activities such as downloading files, forwarding information found on the Web can at times be violations of intellectual property laws. Other activities such as making copies of information for private use may require difficult interpretation of fair use provisions of the law to simply justify their legality. Consequently, individuals in their daily lives have the capability and the opportunity to access and copy vast amounts of digital information, yet lack a clear picture of what is acceptable or legal. On the other hand, the necessary amendments of legislation in several cases do not fully cope with the problem, resulting in certain legislative weaknesses.
   
   
3. A fundamental problem is that institutions in the Cultural Heritage sector want to make information widely available for educational, non-commercial reasons, but the legal environment makes this difficult. Where the rights-holders are known, this can be negotiated, but the costs of clearing the rights for images taken by an individual for non-commercial purposes can be prohibitively expensive.
   
   

It is the aim of this work to present current technological solutions for IPR protection, which are designed, implemented and applied through cultural applications, European projects, etc., and to discuss recommendations, key points, disputes and limitations of these technological solutions.

Technological Solutions Overview

The technical part of the IPR protection of digital material problem focuses on how to provide access without giving up control. A complete technological schema which incorporates all possible means for IPR protection includes:

• Technical Protection Means (TPM): that is a technology that supports users, content owners and organizations to secure and protect digital content (text, image, video, sound, graphics) from unauthorized use. The definition implies the traceability of an improper use. Proposed TPM are:
  • Security and integrity of OS and computer networks.
  • Encryption of transferred data.
  • Watermarking of multimedia content.
  • Tracking the use of the protected content.
  
  
• Digital Rights Management Systems (DRMS): that is systems supporting the management of rights of digital content for providers and users including time and usage based business models.
  • Identification Systems.
  • International metadata standards for IPR management.
  • Rights management programming languages
  • Formats
  • Delivery methods and technologies.

The above generic schema is applied, amongst other, to the Cultural sector and provides the basis of an efficient IPR protection and management of digital media.

In some papers and practices the term Digital Rights Management includes also the Technical Protection Means. In these cases a complete DRMS includes also watermarking software, data hiding technologies and encryption. In this paper, Technical Protection Means are considered as independent and stand - alone applications which serve as a platform mainly for digital content protection from unauthorized use.

Technical Protection Means

The TPM used by practices and projects could be summarized below:

• Security and integrity features of computer operating systems (include, for example, the traditional file access privileges enforced by the system).
  
  
• Encryption, allows digital works to be scrambled so that they can be unscrambled by legitimate users only.
  
  
• Persistent encryption, allows the consumer to use information while the system maintains it in an encrypted form.
  
  
• Watermarking or data hiding, embeds information (e.g. about ownership) into a digital work in much the same way that paper can carry a watermark. A digital watermark can help owners track copying and distribution of digital works.
  
  
• Trusted Systems. In one vision of the future, security will become a major influence on the design of computing infrastructure, leading to the development and widespread adoption of hardware based, end-to-end systems that facilitate control of digital IP. The "trusted systems" constitute an open research area.
  
  
• Protection Technologies for niches and Special - Purpose Devices.

Whether a TPM is successful lies on its technical strength and depends also on both the product it protects and the business in which it is deployed. The most important features are:

• Usability. A difficult to use protection system may discourage users from using it.
  
  
• Appropriateness to the content. The cost of designing, developing, and deploying the system has to be in harmony with the type of the content. For inexpensive or already available in a reasonably priced, non-Internet medium, there is no point to an expensive TPM that drives up the price of Internet delivery.
  
  
• Appropriateness to the threat. Preventing honest customers from giving copies to their friends may require nothing more that a reasonably priced product, a good distribution system, and a clear set of instructions. Preventing theft of extremely valuable content that must at some point reside to a computer network requires a very sophisticated TPM, and even the best available with current technology may not be good enough.
  
  
• The cost-benefit analysis is a difficult but necessary task.

In the following, a more detailed description about data hiding is given, since this technology still appears a novelty for most of the people working in the Cultural Heritage sector.

3.1 Data hiding technologies

Digital watermarking or data hiding is a multidisciplinary research field that combines signal processing with cryptography, communication and coding theory. More specifically, by data hiding we mean the concealment of some information, also called digital watermark, within a host digital media (digital images, sound, video and computer graphics); the information is hidden within the data itself, instead of being attached to the data as a header or as a separate file.
Concealment is achieved by modifying some features of the host media in a way that the modifications are not perceived by a user, so that the quality of the host data is not degraded by the embedding process and also because in this case is it is more difficult to remove the hidden data. For example, in the case of a digital image, it is possible to embed the information into the spatial domain, by properly changing some pixel colour values, or into a transformed domain: in this case, a mathematical transformation (like a Discrete Wavelet Transform or a Discrete Cosine Transform) is applied to the image obtaining a new representation of the image itself; next, some Transform coefficients are properly modified, and finally the inverse transformation is carried out to obtain the modified image. The casted information can be extracted afterwards from the modified content by detecting with a proper decoder the modifications introduced by the data hiding system.

Data hiding techniques were first applied to copyright protection applications, where the embedded signal, i.e. the watermark, conveys copyright-related information about the hosting data. The exact content of the embedded information depends on the particular application and may include the identity of the creator or the distributor of the work, or of the customer whom the work was sold, or the licensing terms between seller and purchaser. The casted information can be used to demonstrate content ownership, content misappropriation or as a proof of purchase, and to do so, these data must remain intact after possible modifications of the work. In such an application, it is required that the digital watermarking is robust (i.e. it is resistant) to intentional or unintentional modifications of the watermarked content carried out during its fruition, that could delete of make unreadable the embedded information.

Later on, new application fields demonstrated to take advantage of the use of data hiding: e. g. this technology can be useful to verify whether the content has been modified since its distribution. In such a case, a pattern, i.e. the watermark, is embedded into digital data to verify whether the content (i.e. an image) has been modified or falsified since its creation. The inserted data are extracted from the possibly corrupted object and compared with the original embedded code: if they match then the content is assumed to be uncorrupted; otherwise, if the extracted code does not match the embedded one, it means that tampering occurred. In this application, it is required that the digital watermark is fragile (i.e. not resistant) to modifications carried out to the watermarked content.

Data hiding can also be used to embed hidden labels and annotations into the host data: in this case, the embedded code contains an annotation with information about the content; for example, a digital image of some Cultural Heritage could embed some information like an Identification Image Number, the content owner, the time the picture was taken, etc. Also in this case, not to lose the embedded data the system must result robust against possible unintentional manipulations of the image.